Comet
PulseAugur coverage of Comet — every cluster mentioning Comet across labs, papers, and developer communities, ranked by signal.
- developed by Perplexity AI 100%
- acquired by Atlas Ai Model 90%
- instance of Atlas Ai Model 90%
- competes with Safari 80%
- competes with Brave 70%
- instance of Brave 70%
- competes with Detroit Institute of Arts 70%
- used by Atlas Ai Model 70%
- competes with opera 70%
- used by United States Court of Appeals for the Ninth Circuit 70%
- affiliated with United States Court of Appeals for the Ninth Circuit 70%
- located in United States Court of Appeals for the Ninth Circuit 70%
- 2026-05-24 research_milestone Researchers discovered and disclosed a prompt injection vulnerability in the Perplexity Comet browser. source
5 day(s) with sentiment data
-
CoMET method improves 3D vehicle detection by tracking roadside pose errors
Researchers have developed CoMET, a novel method for improving asynchronous vehicle-infrastructure cooperative 3D detection. CoMET addresses the challenge of pose errors in roadside sensor messages, which can shift dete…
-
LLMs evaluated for English-Arabic dialectal translation quality
A new research paper evaluates the performance of various large language models (LLMs) in translating between English and dialectal Arabic across 16 different dialects. The study utilized the MADAR dataset and employed …
-
AI agents spark legal fights over value capture for consumers and workers
A legal battle is unfolding over the value generated by AI agents, with implications for consumers, workers, and citizens. In one case, Amazon sued Perplexity over its AI agent browsing its site, arguing unauthorized ac…
-
FlowTranslator offers training-free in-image translation using diffusion models
Researchers have developed FlowTranslator, a novel framework for in-image machine translation (IIMT) that does not require paired training data. This method leverages advanced text-to-image diffusion models by first usi…
-
FlowTranslator enables training-free in-image machine translation
FlowTranslator is a novel framework that enables training-free in-image machine translation by leveraging diffusion models and flow-based editing techniques. This approach generates accurate translations within images w…
-
Walmart, Amazon diverge on AI shopping agent integration · 3 sources tracked
Retailers like Walmart and Amazon are developing distinct strategies for integrating AI shopping agents into their platforms. Walmart plans to allow third-party agents with spending limits and profile management, while …
-
New COMET framework enhances AI safety monitoring with multi-layer coordination
Researchers have developed COMET, a novel multi-layer runtime monitoring framework designed for AI systems used in safety-critical applications. This framework aims to provide human supervisors with actionable insights …
-
Offline speech translation system self-corrects on edge devices
Researchers have developed an offline speech-to-speech translation system that can run on a Jetson Nano edge device and self-correct its translations without needing retraining. The system utilizes a Whisper-tiny ASR mo…
-
COMET metric shows script bias in translation evaluation, new method proposed
A new research paper identifies a significant script bias in the COMET metric, which is used to evaluate machine translation quality. The study demonstrates that COMET scores can vary substantially based on the script u…
-
Retailers Divided on AI Shopping Agents: Hard Blocks vs. Nuanced Gates
Retailers are grappling with the integration of AI shopping agents, with a significant divide emerging on how to handle them. While AI excels at understanding customer intent, the underlying commerce platforms must stil…
-
AI tutors show promise in resolving student impasses and matching human learning gains · 3 sources tracked
Recent research explores the effectiveness of AI tutors in educational settings, particularly in resolving student impasses. One study analyzed over 20,000 student interactions with an LLM chemistry tutor, identifying c…
-
New EviSI agent improves evaluation of simultaneous translation
Researchers have developed EviSI, a new evaluation agent designed for simultaneous speech-to-speech translation systems. Unlike traditional metrics like BLEU and COMET, EviSI incorporates Multidimensional Quality Metric…
-
Ninth Circuit rules for Perplexity AI, freeing its shopping agent from Amazon's claim · 5 sources tracked
The Ninth Circuit Court of Appeals has ruled in favor of Perplexity AI, vacating an injunction that Amazon.com Services, LLC had sought to block Perplexity's AI shopping agent, Comet. The court determined that it is the…
-
Google Chrome accelerates release cycle to bi-weekly amid AI security shifts · 3 sources tracked
Google Chrome has officially transitioned to a bi-weekly release schedule, moving from its previous four-week cycle. This acceleration is driven by the evolving security landscape influenced by AI, enabling faster deplo…
-
Perplexity's Sonar API is key for autonomous research, but costs extra
A user explored integrating Perplexity into their deep-research workflow, concluding that its Sonar API is the only viable option for autonomous, unattended research tasks. This API is billed separately from Perplexity …
-
New LLM agents evaluate simultaneous interpreting, improving on BLEU and COMET
Researchers have developed new methods for evaluating simultaneous interpreting systems, addressing limitations of existing metrics like BLEU and COMET. One approach, detailed in an arXiv paper, uses a LoRA-adapted COME…
-
Gemma 2 and 3 translation features show limited cross-lingual transfer
A new research paper investigates the cross-lingual validity of Sparse Autoencoder (SAE) features in Google's Gemma 2 and Gemma 3 language models. The study found that while many features appear frequently across differ…
-
MLflow faces competition from 7 alternatives for ML experiment tracking
MLflow, a popular tool for ML experiment tracking since its 1.0 release in 2018, is now facing competition from several alternatives. The article highlights seven such tools that offer similar or enhanced capabilities f…
-
New benchmark evaluates machine translation for Chinese social media slang
Researchers have developed CSM-MTBench, a new benchmark designed to evaluate machine translation (MT) systems specifically for Chinese social media texts. This benchmark addresses challenges like the rapid evolution of …
-
SemEnrich method enhances radiology report datasets for vision-language learning
Researchers have developed SemEnrich, a self-supervised method to improve vision-language learning datasets for radiology reports. This technique uses semantic clustering to enrich reports with positive or neutral obser…